RADV audits, risk-adjustment FCA matters, and contractor audits

Per-submitted-code documentation review for RADV audits, risk-adjustment FCA matters, and OIG, UPIC, and RAC contractor audits — verbatim evidence, and a person's signature on every finding.

RADV Audit Response

In a Risk Adjustment Data Validation (RADV) audit, the question for each audited diagnosis is whether the medical record supports it. Someone still has to answer that question one code at a time — is this diagnosis supported by what the clinician documented?

  • Every submitted diagnosis in the audit sample that carries risk-adjustment weight read against its note, with the exact line cited where a read rests on it — including negation, resolution, and history-of language.
  • Whatever note evidence the software finds is pulled for every code in scope, so the reviewer's hours go to deciding the calls, not to searching for the note language behind them.
  • Relative risk-adjustment weight surfaced only as a triage input — never a dollar figure, never a debt total.
  • Detected uncertainty resolves to review, never to "supported" — and because the software can still misread a code as supported without detecting any doubt, every code it reads as supported goes to a person.

What we are. A per-code chart review, accelerated by software and decided by a person. A person, not the software, reviews every code and signs the findings; your engagement letter names who that is. Our coding lead approves the clinical criteria the software applies before any version of them is used on a matter. She holds no coding certification today. The FAQ says what that means for your engagement.

What the AI does. Two different AI models each read the note for every code in the sample that carries risk-adjustment weight, separately, and judge from the whole sentence whether the condition is present at this visit or is ruled out, resolved or historical. Neither model decides: fixed rules read a code as supported only when both readings agree and every check passes. A final automated check can send a code to review, never the other way; it runs on the same model as the second reading, so it is an internal check, not an independent one. Whatever the read, every code then goes to a person. The software does not extrapolate a sample finding across the contract. If the audit reaches that stage, that opinion belongs to a statistician.

Request a Call

One RADV notice. One question to answer, per code: is this diagnosis supported by the record?

Risk-Adjustment FCA & DOJ Matters

A risk-adjustment False Claims Act matter — a qui tam complaint from a data-mining relator, or a DOJ civil investigative demand — can reach several years of diagnosis submissions at once. Before counsel can assess exposure or answer a CID, someone has to work through that lookback period the way a RADV auditor would: one submitted diagnosis, one chart, one question.

  • Every submitted diagnosis named in the CID or the complaint's lookback period that carries risk-adjustment weight read against its note, with the verbatim language cited where a read rests on it.
  • Findings organized the way a data-miner's complaint is organized — one submitted code at a time. For most codes the software's best read is "needs review," so the answer on what the record shows comes from the reviewer, not the software.
  • Retained under your engagement as a consulting expert, not a testifying one, supporting privilege through discovery.
  • Where the software detects doubt, the code routes to review — it never resolves doubt in the client's favor. It can still misread a code as supported without detecting any doubt, which is why every code it reads as supported also goes to a person.

The same structure and the same two-model read described under RADV apply here: a person — not the software — reviews every code and signs the findings. Our coding lead, who approves the clinical criteria the software applies, holds no coding certification today, and the FAQ says what that means for your engagement. The software does not compute a dollar total of any kind — see what we will not sign — and it does not adjudicate; the finding is evidence for counsel's judgment, not a verdict.

Request a Call

A qui tam complaint or a DOJ civil investigative demand spanning years of risk-adjustment submissions. The question is still one diagnosis at a time: does the record support it?

OIG, UPIC & RAC Contractor Audits

Where an OIG, UPIC or RAC audit turns on whether the record supports a submitted diagnosis code, the question is the same one, answered one code at a time: does the record support it?

  • Every submitted diagnosis in the audited code group that carries risk-adjustment weight read against its note, with the verbatim language cited where a read rests on it.
  • We do not perform sampling or extrapolation ourselves. That opinion belongs to a statistician.
  • The agreement/disagreement record your team needs when weighing whether to contest, self-disclose, or negotiate a contractor's findings.

Posture stays the same across every matter type on this page: retained by you, under your engagement, as a consulting expert — not a testifying one.

Request a Call

Where an OIG, UPIC or RAC audit turns on a submitted diagnosis code, the question is the same, per code: is this diagnosis supported by the record?

What You Receive

Across every matter type above, the deliverable is the same shape: a signed work product, not a platform login.

  • The full matter run — every submitted diagnosis that carries risk-adjustment weight read against its note, with the note language pulled and cited where a read rests on it, and every code with no payment category listed as out of scope.
  • The signed review queue — the reviewer's findings, including where the reviewer's decision differed from the software's read.
  • The agreement/disagreement record — the directional review record showing what the reviewer examined and where the reviewer disagreed with the software's read.
  • Relative risk-adjustment weight, shown only as a triage input so your team knows what to look at first. There is no dollar column on the deliverable.

What's Not Included

  • Statistical sampling and extrapolation. That opinion belongs to a statistician, whose declaration we may not sign — if a matter needs it, we say so and stay in our lane.
  • Adjudication or any dollar total. We do not compute a debt or a damages figure of any kind.
  • Testifying-expert work. We are retained in a consulting-expert lane by design.
See how engagements are structured →